On Thu, Nov 14, 2013 at 01:45:29AM +1100, Russell Coker wrote: > Nov 14 00:00:01 workstation /USR/SBIN/CRON[30993]: (root) CMD > (/usr/local/sbin/btrfs-make-snapshot minutes /home >> /var/log/snapshot.log) > > > > Nov 14 00:00:01 workstation /USR/SBIN/CRON[30994]: (root) CMD > (/usr/local/sbin/btrfs-make-snapshot days /home >> /var/log/snapshot.log) > > > > Nov 14 00:00:01 workstation /USR/SBIN/CRON[30995]: (root) CMD > (/usr/local/sbin/btrfs-remove-snapshots 100 100 /home/ / >> > /var/log/snapshot.log) > > Below is a grep for NMI in my kernel message log. I'm running Debian kernel > 3.11.6-2 on a quad-core AMD64 system. Above is the relevant section of the > cron log. It appears that running two scripts that create snapshots at the > same time as a script that removes maybe 97 snapshots causes an NMI. I've > attached the kernel message log from today, I can give you logs from other > days if they are useful.
Snapshot deletion may touch a lot of data and some operations are slowed down by that. This may trigger the warning INFO: task T blocked for more than 120 seconds If there's no other crash (that would block the system indefinitely), the warnings disappear. According to the grep output, the warnings happen at the same daytime, so this is likely the system overload reason. You may want to follow the warning's sugeestion: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message or increase the timeout. david -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
